In spite of the approximately 50 morphological and physiological criteria currently used in yeast systematics, DNA base composition studies have shown a wide intraspecific variation in moles per cent guanine plus cytosine in the nuclear DNA of a number of presently accepted yeast "species". Conversely, the indiscriminate use of a large number of such criteria, has led to the description of many new "species" sometimes differing by only a single biochemical ability, for example hydrolysis of a disaccharide. In several of the latter cases we have shown high degrees of DNA-DNA homology between species separated on this basis. Such "species" may differ by only a single gene or even by a single base pair in the case of point mutations. Only sparse information is available on phylogenetic relationships in yeasts. Support for primitive or advanced species or genera might be deduced from data or genome sizes. For these reasons we propose to study a) additional numbers of strains of certain species to determine variation in DNA base composition; b) DNA-DNA homology between nuclear DNAs of what appear to be closely related species with similar base composition; c) determine genome sizes and per cent of repetitive DNA of selected species; d) DNA-DNA homology between mitochondrial DNAs of species in a number of genera. These approaches are expected to result in a more rational system of yeast classification and to provide clues on phylogenetic development because entire or nearly entire genomes will be compared.